3
$\begingroup$

$\newcommand{\ord}{\text{ord}}$This is a question in my book:

$G$ is an infinite Abelian group and $H=\{ a \in G \mid \ord(a) < \infty \}$ is a normal subgroup of $G$, show that $eH$ is the only element that has an finite order.

And this is my attempt:

$eH=H$ and $aH=H$ for all $a \in G$ with $\ord(a) < \infty$, because then $a \in H$. And we know $eH$ has $\ord(eH)=1$ so that is the only element with a finite order. If we take $a \in G$ and $a \notin H$, then $\ord(aH)$ is infinite. I hope this is correct!

$\endgroup$
  • $\begingroup$ you proved nothing, in fact. Read the answer below: it is clear and to the point. $\endgroup$ – DonAntonio Jun 6 '14 at 14:13
4
$\begingroup$

If $aH$ has finite order in $G/H$, then $(aH)^n=H$, for some $n$. This means that $a^n\in H$. By definition of $H$, there is $m$ such that $(a^n)^m=1$. Hence $a^{mn}=1$ and so $a\in H$ and $aH=H=eH$.

Two remarks:

  • $H$ is called the torsion subgroup of $G$ and denoted $tor(G)$. The result says that $G/tor(G)$ is torsion-free.

  • The hypothesis that $G$ is infinite is not needed. The hypothesis that $G$ is abelian is crucial because it ensures that $H$ is a subgroup (and normal). If $G$ is not abelian, $H$ will only be a normal subset.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.